Identification of Anonymous MRI Research Participants with Face-Recognition Software
Open accessAbstract
Facial Identification from Cranial MRI With the use of publicly available software, reconstructed facial images from deidentified cranial MRI scans were matched to photographs of individual study p...
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The same methods that enable new analyses of clinical data can also allow patient re-identification in ways that are sometimes unpredictable.
“Furthermore, the same methods that are enabling new analyses of clinical data may also enable the re-identification of patients in sometimes unpredictable ways17.”
From Minimum information about clinical artificial intelligence modeling: the MI-CLAIM checklist. (2020) · cites this paper as [17]
The paper reports that face-recognition software matched photographs to reconstructed cranial MRI faces for 70 of 84 participants, demonstrating a potential route to re-identification. It does not establish that the same methods enable new analyses of clinical data or describe the risk as unpredictable. (LLM, conf=0.82)
Experimental — accuracy not yet measuredRemoving metadata such as patient name or date of birth is often not enough to preserve privacy, and it is possible to reconstruct a patient's face from CT or MRI data.
“Even if data anonymisation could bypass these limitations, it is now well understood that removing metadata such as patient name or date of birth is often not enough to preserve privacy7.”
From The future of digital health with federated learning (2020) · cites this paper as [8]
Re-identification attacks have been shown to yield high success rates both with tabular data and medical imaging data.
“Such re-identification attacks 39 have been shown to yield high success rates both with tabular data 40,41 (such as patient records) and medical imaging data 42 .”
From Secure, privacy-preserving and federated machine learning in medical imaging (2020) · cites this paper as [42]
Three-dimensional medical imaging can be volumetrically rendered to reconstruct facial contours which enable patient re-identification.
“three-dimensional medical imaging can be volumetrically rendered to reconstruct facial contours which enable patient re-identification9.”
From Medical imaging deep learning with differential privacy (2021) · cites this paper as [9]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 8 · Cited by 215
- Authors
- Schwarz, C. G., Kremers, W. K., Therneau, T. M., Sharp, R. R., Gunter, J. L., Vemuri, P., Arani, A., Spychalla, A. J., Kantarci, K., Knopman, D. S., Petersen, R. C., & Jack, C. R.
- Affiliations
- Mayo Clinic (US)
- ORCID
- Venue
- New England Journal of Medicine
- Publisher
- Massachusetts Medical Society
- Published
- 2019-10-24
- Volume
- 381
- Issue
- 17
- Pages
- 1684-1686
- ISSN
- 0028-4793 (ISSN-L)
- 1533-4406
- PMC
- PMC7091256
- OpenAlex
- W2982036758
- Semantic Scholar
- e9b8a10d1a2dbc3506ec9aa5d45341b20d547737
- Cited by
- 215
- References
- 8
- Citations by year
- 2026: 27 · 2025: 32 · 2024: 35 · 2023: 29 · 2022: 39 · 2021: 30 · 2020: 22 · 2019: 1
- FWCI
- 10.39
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- New England Journal of Medicine (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- The Alexander Family Alzheimer’s Disease Research Professorship of the Mayo Clinic
- The Liston Award
- The Schuler Foundation
- The Mayo Foundation for Medical Education and Research
- National Institute on Aging (AG006786, AG011378, AG016574, AG034676, AG041851, AG056366)
- National Institute of Neurological Disorders and Stroke (NS097495, NS100620)
- The GHR Foundation
- The Elsie and Marvin Dekelboum Family Foundation
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Face recognition and analysis
- Life Sciences › Neuroscience › Cognitive Neuroscience › Face Recognition and Perception
- Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Ethics in Clinical Research
- Keywords
- face recognition, facial reconstruction, participant identification, research participant privacy, brain MRI
- MeSH terms
- Facial Recognition, Humans, Magnetic Resonance Imaging, Pattern Recognition, Automated, Software, Privacy, Research Subjects
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
